Skip to main content
POST
/
voice
/
text_to_audio
Generate audio from text with voice cloning
curl --request POST \
  --url https://modelslab.com/api/v6/voice/text_to_audio \
  --header 'Content-Type: application/json' \
  --data '
{
  "key": "<string>",
  "prompt": "<string>",
  "init_audio": "<string>",
  "voice_id": "<string>",
  "language": "english",
  "speed": 1,
  "base64": false,
  "temp": false,
  "stream": true,
  "webhook": "<string>",
  "track_id": 123
}
'
{
  "status": "success",
  "generationTime": 123,
  "id": 123,
  "output": [
    "<string>"
  ],
  "proxy_links": [
    "<string>"
  ],
  "future_links": [
    "<string>"
  ],
  "links": [
    "<string>"
  ],
  "meta": {},
  "eta": 123,
  "message": "<string>",
  "tip": "<string>",
  "fetch_result": "<string>",
  "audio_time": 123
}

​
Request

Make a POST request to below endpoint and pass the required parameters as a request body.
Note: You can either pass init_audio or voice_id. However, if both are passed at the same time the init_audio takes preference.
curl
--request POST 'https://modelslab.com/api/v6/voice/text_to_audio' \

​
Language-Specific Guidelines

When using the Text-to-Audio API for Hindi language generation, follow these best practices for accurate and natural-sounding output.
Best Practices for Hindi Language
  • Use clear and well-structured Hindi sentences.
  • Ensure proper punctuation for better clarity.
  • Write numbers in Hindi text format instead of digits.
  • Correct: “दो हज़ार पच्चीस”
  • Incorrect: 2025
  • Correct: “पंद्रह अगस्त उन्नीस सौ सैंतालीस”
  • Incorrect: 15/08/1947
  • Always use expanded Hindi form for dates and numbers.
  • Convert English abbreviations to Hindi phonetic spelling.
  • Correct: आईआईटी (for IIT)
  • Correct: यूएसए (for USA)
  • Correct: एमएल (for ML)
  • Spell out abbreviations in Hindi script to improve pronunciation.

​
Body

json
{    
  "key": "your_api_key", 
  "prompt":"Narrative voices capable of pronouncing terminologies & acronyms in training and ai learning materials.", 
  "init_audio":"https://pub-f3505056e06f40d6990886c8e14102b2.r2.dev/audio/tom_hanks_1.wav", 
  "language":"english", 
  "webhook": null, 
  "track_id": null
}

Body

application/json
​
key
string
required

API key for authentication

​
prompt
string
required

Text prompt describing audio to be generated

​
init_audio
string<uri>

Valid URL pointing to audio file for voice cloning (4-30 seconds)

​
voice_id
string

ID of voice from available list Find Voice IDs Here

​
language
enum<string>
default:english

Language for the voice

Available options:
english,
arabic,
spanish,
brazilian portuguese,
german,
czech,
chinese,
dutch,
french,
hindi,
hungarian,
italian,
japanese,
korean,
polish,
russian,
turkish
​
speed
number
default:1

Playback speed of generated audio

​
base64
boolean
default:false

Whether input audio is in base64 format

​
temp
boolean
default:false

Use temporary links valid for 24 hours

​
stream
boolean

Stream response in base64 format

​
webhook
string<uri>

URL to receive POST notification upon completion

​
track_id
integer

ID for webhook identification

Response

Text to audio response

​
status
enum<string>

Status of the voice generation

Available options:
success,
processing,
error
​
generationTime
number

Time taken to generate the audio in seconds

​
id
integer

Unique identifier for the voice generation

​
output
string<uri>[]

Array of generated audio URLs

​
proxy_links
string<uri>[]

Array of proxy audio URLs

​
future_links
string<uri>[]

Array of future audio URLs for queued requests

​
links
string<uri>[]

Array of audio URLs (voice cover response)

​
meta
object

Metadata about the audio generation including all parameters used

​
eta
integer

Estimated time for completion in seconds (processing status)

​
message
string

Status message or additional information

​
tip
string

Additional information or tips for the user

​
fetch_result
string<uri>

URL to fetch the result when processing

​
audio_time
number

Duration of the generated audio in seconds